Support Image2DICOM Conversion

[tostein]

@fedorov As discussed on the NAMIC project week 30, the need of converting non-DICOM image data (ITK formats) into DICOM is present. An example use case are data provided by challenges. We would like to have this functionality in MITK, but if this feature would be included in dcmqi others can profit as well

I think the mechanism of dcmqi (providing meta data as JSON) fits perfect to supply a converter taking the image pixel data and fills out the necessary DICOM tags. In any case, it would be nice to reuse some of the existing code to pursue this goal.

It would be great to have such a converter within the dcmqi library. If this idea finds consent i would be interested to start the development.

[fedorov]

@tostein it would be wonderful to have such functionality in dcmqi, and your contribution is most welcomed!

I think the starting point should be the analysis of this post on the DICOM4QI forum from @dclunie.

Do you want to take a look at that and maybe we could schedule a phone call to discuss the next steps?

[fedorov]

@tostein @nolden do you have any thoughts? Is this still of interest to investigate to you?

[nolden]

https://github.com/tostein/dcmqi/tree/image2dicom

[fedorov]

Per discussion today:

• tool for conversion from single-frame to legacy-converted enhanced MR: https://github.com/QIICR/enhanced_mr (I will add @tostein to that repo)
• unfinished PR refactoring dcmqi: WIP: refactoring of the code base #226
• we use dcmGenerateUniqueIdentifier() function to generate UIDs

[fedorov]

I added some comments to the branch. Overall, it looks consistent with the rest of dcmqi to me, the question is how the next steps will be implemented.